Xander S.

atexit feels like the right way to handle this. Documentation on how to handle this could be helpful, though.

Hey Ayub K., we don't have affordances for that in the UI. You should be able to accomplish this using the client by pulling and combing the various datasets. https://arize-phoenix.readthedocs.io/projects/client/en/latest/api/datasets.html

I take it the session IDs that are colliding are coming from different teams who don't know what session IDs the other teams are using?

I see, thanks for the context. The easiest unblock for now then is to ensure that the session ID is globally unique. How are those session IDs being generated now?

We have an assumption in our data model that a session belongs to exactly one project, which is the project associated with the first trace of the session that Phoenix receives. We assumed when building this feature that session IDs would contain a random element to them (so collisions are unlikely) and that users would typically send the traces from the same session to a single project. https://github.com/Arize-ai/phoenix/blob/main/src/phoenix/db/README.md#core-tracing--projects

After sending the second request, do you see your traces from the second request in the session in the first project?

Can you show me the code?

The trace goes to the correct project, but the session grouping happens globally based on session_id.

Hey Brais M., what session grouping are you referring to? Are you referring to the Sessions tab? Maybe you could send a screenshot.

Thanks Nas. It's on our radar. For now, I would make a second message or message content part to include the reasoning. https://github.com/Arize-ai/openinference/issues/1649

This is our legacy experiments API, so might be worth trying the new client to see if the issues persists Nico G..